FASSTEST

Accessibility statement
FASS Centre for Scholarship and Innovation logo

You are here

  1. Home
  2. Integrating OUAnalyse into DE100 to Aid Retention and Progression

Integrating OUAnalyse into DE100 to Aid Retention and Progression

OUAnalyse predicts on a weekly basis whether, or not, a student will submit their next assignment. The dashboard visualises predictive information about who is at risk of not submitting their next assignment, as well as module website engagement, and assignment submission rates at the cohort level. It does this through a traffic light system. These predictions are made using machine learning algorithms utilising two types of data: (a) static data: demographics such as age, gender, geographic region, previous education, and (b) behavioural data: students’ interactions within the module website. These sources of data have been shown to be significant indicators of predicting students’ assignment submission (e.g., Kuzilek et al., 2015). 

Previous research has indicated that using OUAnalyse can positively influence students’ performance.  Specifically, the results of Hlosta, Herodotou, Fernandez, and Bayer (2021) demonstrate that this positive impact was particularly evident in those who were coming from low SES, as measured by the Index of Multiple Deprivation (IMD). BAME students are shown to have the greatest representation in low SES (32% as opposed to 10% non-BAME students), stressing the importance of predictive analytics in supporting BAME students. Because this study was conducted only using three modules it is important that to see if this effect can be replicated in different modules. 

The OUAnalyse team have provided various useful training information, however, due to time constraints we believe tutors might find it difficult to know how to implement strategies whilst tutoring.  Therefore, we believe that tutors would benefit from tailored, simplified module specific training on how to use dashboard. But critically also being provided with a set of actions that include templated emails aimed at supporting their students. By creating this framework, we are offering a structured approach to using the tool on a regular basis. 

We know tutors are interested in using the dashboard due to the high number of attendees at the Psychology conference session.  During the session, it became apparent that tutors needed clear guidance on how the dashboard could be used for specific modules in a targeted way, which was the impetus for this project. 

Through the development of a tutor Training Package that offers a framework of appropriate contact points and actions, with supporting resources (such as email proformas) we aim to capture and support students at vulnerable points through an observation of their online module engagement data. 

This project will: 

  1. Create a module-specific Dashboard Framework to support tutors 

  1. Create a training programme for tutors 

  1. Evaluate the Dashboard Framework with respect to student retention and ease of use 

  1. Based on the feedback explore how subsequent Dashboard Frameworks could be designed for other modules  

Relevant literature: 

Herodotou et al. (2020) ‘How Can Predictive Learning Analytics and Motivational Interventions Increase Student Retention and Enhance Administrative Support in Distance Education?’ doi: 10.18608/jla.2020.72.4. 

Hlosta, M., Herodotou, C., Fernandez, M., and Bayer, V (2021). Impact of Predictive Learning Analytics on Course Awarding Gap of Disadvantaged students in STEM. In: 22nd International Conference on Artificial Intelligence in Education, AIED 2021, Lecture Notes in Artificial Intelligence, Springer. 

Kuzilek, J., Hlosta, M., & Zdenek, M. (2017) Open University Learning Analytics dataset. Scientific Data 4:170171 doi: 10.1038/sdata.2017.171. 

Rienties et al. (2016) ‘Reviewing three case-studies of learning analytics interventions at the open university UK’, in. ACM. doi: 10.1145/2883851.2883886.